医疗器械临床试验单组目标值设定现状

Current Status of Performance Goals Setting in Single Arm Medical Device Clinical Trials

单组目标值法是临床研究设计的重要类型之一,特别是在医疗器械的临床试验领域。我国和美国等国家和地区的医疗器械监管机构制定的指南等相关文件都对单组目标值法进行了论述,但尚未明确提出目标值具体该如何设定。本文分析了近年来采用单组目标值法设计的48篇医疗器械临床试验文献,结果显示当前目标值的设定主要以“参考值+界值”的方法为主,目标值相对参考值的比值范围较大,从1.10~11.00,中位数为1.50(IQR 1.38~2.06),且随着参考值的增加而减少。因此,笔者建议在医疗器械的临床研究设计中当参考值<1%,1%~<5%,5%~<10%,≥10%时,目标值分别不宜超过参考值的3倍、2倍、1.75倍、1.5倍,如超过该范围应当有充分的理由。

Performance Goals(PGs)in single-arm studies are important in clinical study design,especially in medical device clinical trials.Medical device administrations in countries like China and the US have issued documents introducing PGs;however,specific ways to set a PG are unidentified.In this article,we reviewed 48 medical device clinical trials using PGs in single-arm studies in recent years,and found that the majority adopted the method of using a reference value and a critical value to determine a PG,with a varied ratio of the PG to the reference value,ranging from 1.10 to 11.00(median 1.50,IQR 1.38~2.06),and the ratio decreasing as the reference value increased.Thus,we suggest that when the reference value is<1%,1%~<5%,5%~<10%,≥10%,the PG should not exceed 3,2,1.75 or 1.5 times the reference in medical device clinical study designs;otherwise,justification should be stated.